-
公开(公告)号：US20220156514A1
公开(公告)日：2022-05-19
申请号：US17454743
申请日：2021-11-12
Applicant: QUALCOMM Technologies, Inc.
Inventor： Kirill GAVRILYUK , Mihir JAIN , Cornelis Gerardus Maria SNOEK
Abstract: Certain aspects of the present disclosure provide techniques for training a first model based on a first labeled video dataset; generating a plurality of action-words based on output generated by the first model processing motion data in videos of an unlabeled video dataset; defining labels for the videos in the unlabeled video dataset based on the generated action-words; and training a second model based on the labels for the videos in the unlabeled video dataset.
Search Results
1
records
(took 0.012 seconds)
筛选
AND
AND
过滤
NOT
NOT
Scan to Join
